I am not talking about players exploiting/hacking the structure, game theory have nothing to do with collusion. I am talking about 2 players (or more) exchanging informations in order to gain an edge, or losing deliberatly against each other etc ...
I am a veteran poker player and well aware of the current centralized approach to this issue; simply police and enforce, casino/gaming platforms have the means to detect and prove any wrongdoing (access to history, machine ids etc) and act as it sees fit to compensate, ban or whatever.
but in a decentralized environnement how are you going to make sure nobody is colluding? and if it happens, who can settle a dispute or a claim ?
@MarketingMonk, I believe the situation
@Nietzsche26 is speaking about requires 2 factors, both of which must be present in order for collusion to be an effective strategy.
First, the game must have multiple players sitting at the same table and playing from the same single deck of cards (as opposed to "head to head" games). Second, the game must include the opportunity for additional rounds of betting after the ante round.
Texas hold'em is a good example that most people can relate to. If you are playing online at buy-in tables/games of hold'em, most websites allow you to choose your table to "sit down" at and play. Most tables have 6 or 8 players maximum, if I remember correctly.
The way hold'em works, everyone at the table is dealt 2 cards from the deck. Then there is a round of betting. Then 3 cards are dealt face up on the table, then there is another round of betting. Then another card is dealt face up, then another round of betting, and so on. The object is to wager (or bluff) based on your confidence that you have the best poker hand which combines the two cards in your hand with any number of the cards that are face up on the table.
At an 8 person table, the whereabouts of at least 16 cards (2 cards held by each player) would be "known" by the players sitting at the table. Any player who could simultaneously view their own hand and the hand of any of the other players would have a statistical advantage over the rest of the players at the table, because they would know the whereabouts of more cards, and thus be able to better calculate and deduce the odds. There are also ways to "muscle" other players out of the hand by using sophisticated cooperative bluffing strategies.
This obviously cannot be done in live, real world hold'em tournaments, unless you established some sort of elaborate way of signaling with your cheating partners. But it can, and often does happen in online tournaments, where players can hide behind multiple account names and simultaneously play all the accounts at the same table. They can then sucker the "outsiders" who don't realize they are playing at an 8 player table where 2 or more of the "people" are actually the same person with multiple accounts.
Now, while this collusion strategy has worked for cheaters of online Texas hold'em, it would not work in games where there is only the "ante" round of betting. This is because the game would necessarily be "fated" to rely upon whatever hand chance dealt you. While "ante only" games are not really in line with the spirit of Poker, there are plenty of other games that the ante round is the only wager (think beyond card games). Also, this same collusion strategy would not work for two player "head-to-head" games, because if you knew both hands you would just be playing against yourself for your own money.
For most of the initial games on Peerplays, we are implementing a head-to-head, single elimination tournament bracket structure. However, our randomized player matching algorithm is one of the solutions we are working on to greatly eliminate the possibility of collusion in multiplayer buy-in tables for games like Texas hold'em. We will be discussing this more publicly in the months ahead.
Show Posts
